Re: Add unroll-and-jam pass v2

2017-12-07 Thread Richard Biener
On Thu, Dec 7, 2017 at 3:11 PM, Michael Matz wrote: > Hi, > > On Mon, 20 Nov 2017, Richard Biener wrote: > >> > +static void >> > +fix_loop_structure (struct loop *loop, struct loop *old) >> >> We have a function with the same name in loop-init.c so please use a >> different name. > > Renamed to m

Re: Add unroll-and-jam pass v2

2017-12-07 Thread Michael Matz
Hi, On Mon, 20 Nov 2017, Richard Biener wrote: > > +static void > > +fix_loop_structure (struct loop *loop, struct loop *old) > > We have a function with the same name in loop-init.c so please use a > different name. Renamed to merge_loop_tree. > > + for (gsi = gsi_start_bb (bb); !gsi_end_p

Re: Add unroll-and-jam pass v2

2017-11-20 Thread Richard Biener
On Fri, Nov 17, 2017 at 2:58 PM, Michael Matz wrote: > Hi, > > so I've dusted off and improved the implementation of unroll-and-jam from > last year. The changes relative to last submission are: > > * corrected feasibility of the transform (i.e. that dependency directions > are correctly retain

Add unroll-and-jam pass v2

2017-11-17 Thread Michael Matz
Hi, so I've dusted off and improved the implementation of unroll-and-jam from last year. The changes relative to last submission are: * corrected feasibility of the transform (i.e. that dependency directions are correctly retained, the last submission was wrong). * added profitability analys